Despite their best intentions, executives fall prey to cognitive and organizational biases that get in the way of good decision making. In this series, we highlight some of them and offer effective ways to respond.
Our topic this time?
Being objective about budgets
Bias busters: Being objective about budgets
It has been another long, exhausting budget meeting. You reviewed the presentations, challenged every number, explored every assumption. In the end, you raised targets a little, but, if you’re honest, you have to admit it: next year’s targets are not very different from the ones the business units proposed at the beginning of the budget process, which, in turn, are not very different from the latest forecasts for this year. What happened?
You were likely the victim of anchoring. It’s a psychological phenomenon in which a number sticks in your mind and influences you, even if you think you’re disregarding it, and even when the numbers seem irrelevant. It happens every day—inside and outside the strategy-planning room.
Consider this example: a group of people was asked two questions. First, “Was Gandhi younger or older than nine when he died?” And then, “How old do you think Gandhi was when he died?” A separate group was also asked two questions. First, “Was Gandhi younger or older than 140 when he died?” And then, “How old do you think Gandhi was when he died?” Both reference-point numbers in the questions are ridiculous, of course, but those anchors affected how people responded. The first group said Gandhi died when he was 50, and the second group said he died when he was 67. In reality, Gandhi was 78 when he was assassinated.
In the context of business, anchors can similarly make business-unit leaders believe their plans and investments are changing significantly over time when in fact they remain relatively fixed (exhibit).
An anchor is such a powerful influence that only another anchor can overcome it. Reanchoring replaces the number in your head with one grounded in a different set of facts. To see how it works, imagine you are charged with setting your company’s sales targets for several different regions. You could take the following steps to reanchor yourself:
- Set some fact-based, nonhistorical criteria for determining sales targets. These might include, for instance, market growth over a set period, the company’s current market share, and the number of sales representatives in your company compared with competitors. You don’t have to include every single factor, but you should make sure that objective data can be found to document the criteria you’ve set. History (for example, this year’s sales targets) should not be a factor—it already has enough weight as an anchor.
- Build and calibrate a forecasting model based on these criteria. The goal here is to answer a question: “If you did not know what your sales targets were this year and were relying only on the criteria you defined, what would the targets for next year be?” There are many techniques you can use to answer this question; you could, for instance, perform a simple regression analysis using only a few of the variables you’ve identified. Just remember that you are not trying to make absolute predictions; the model only needs to be directionally correct in most cases. If the model’s output is within 10 percent of historical numbers in two-thirds of sales territories, for instance, you probably have something precise enough.
- Use the model as a second anchor. Now you can use the model’s output to change the dynamics of the target-setting discussion. For instance, a budget meeting likely used to start with, “You are on track to deliver 100 units this year, and you’re aiming for 103 next year, but I am sure you can do better.” Now, you can change the conversation to: “You’re aiming for 103 units, but the model tells me you have the potential to aim for 120—let’s talk.” Of course, this is going to be a longer conversation. But for each sales territory in which the two anchors are far apart, there will likely be one or two where the anchors are very close (that is, incoming targets will be close to the model’s output). These discussions can be expedited, which will save discussion time for the difficult cases.
Variations of this reanchoring approach can be used in any target-setting or resource-allocation process where you want to challenge the status quo. It focuses debate where debate is really needed and helps reduce the inertia that anchoring induces.
It is not a panacea: at the end of the day, you will still have to make tough decisions. But reanchoring will help make difficult conversations considerably more productive.
Parts of this article were adapted from “Is your budget process stuck on last year’s numbers?” by Dan Lovallo and Olivier Sibony, March 2014.